[tracing] Add browser-side tests for memory dumps.

[tracing] Add browser-side tests for memory dumps.

This is a rebase of https://codereview.chromium.org/1064413002. The
`OverlappingRequests` test has been removed.

Committed: https://crrev.com/a4c2539f13510a650d51e65204ea52426979659c
Cr-Commit-Position: refs/heads/master@{#343408}

[Ruud van Asseldonk, 2015-08-14 11:46:45]

ruuda@google.com changed reviewers:
+ primiano@chromium.org

[Primiano Tucci (use gerrit), 2015-08-14 14:01:06]

primiano@chromium.org changed reviewers:
+ dsinclair@chromium.org

[Primiano Tucci (use gerrit), 2015-08-14 14:01:06]

LGTM w/ some minor comments thanks.
+dsinclair for OWNERS stamp

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
File content/browser/tracing/memory_tracing_browsertest.cc (right):

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:24: using testing::Gt;
I think you don't need this anymore

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:25: using
testing::InvokeWithoutArgs;
ditto

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:88: std::string
category_filter = TRACE_DISABLED_BY_DEFAULT("memory-infra");
Please use MemoryDumpManager::kCategoryForTesting (or whatever the right name
is)

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:106:
base::MessageLoop::current()->task_runner(), run_loop.QuitClosure());
I think that base::MessageLoop::current()->task_runner() has been recently
deprecated and the new cool thing is ThreadTaskRunnerHandle::get()

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:140:
BrowserInitiatedDumpSingle) {
BrowserInitiatedDumpSingle -> BrowserInitiatedSingleDump (I know, I think this
is my original fault)

[dsinclair, 2015-08-14 14:03:54]

rs lgtm w/ primiano's nits fixed.

[Ruud van Asseldonk, 2015-08-14 15:47:01]

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
File content/browser/tracing/memory_tracing_browsertest.cc (right):

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:24: using testing::Gt;
On 2015/08/14 14:01:06, Primiano Tucci wrote:
> I think you don't need this anymore

Done.

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:25: using
testing::InvokeWithoutArgs;
On 2015/08/14 14:01:06, Primiano Tucci wrote:
> ditto

Done.

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:88: std::string
category_filter = TRACE_DISABLED_BY_DEFAULT("memory-infra");
On 2015/08/14 14:01:06, Primiano Tucci wrote:
> Please use MemoryDumpManager::kCategoryForTesting (or whatever the right name
> is)

Done.

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:106:
base::MessageLoop::current()->task_runner(), run_loop.QuitClosure());
On 2015/08/14 14:01:06, Primiano Tucci wrote:
> I think that base::MessageLoop::current()->task_runner() has been recently
> deprecated and the new cool thing is ThreadTaskRunnerHandle::get()

Done.

https://codereview.chromium.org/1295753002/diff/1/content/browser/tracing/mem...
content/browser/tracing/memory_tracing_browsertest.cc:140:
BrowserInitiatedDumpSingle) {
On 2015/08/14 14:01:06, Primiano Tucci wrote:
> BrowserInitiatedDumpSingle -> BrowserInitiatedSingleDump (I know, I think this
> is my original fault)

Done.

[Ruud van Asseldonk, 2015-08-14 15:55:05]

The CQ bit was checked by ruuda@google.com

[Ruud van Asseldonk, 2015-08-14 15:55:05]

The patchset sent to the CQ was uploaded after l-g-t-m from
primiano@chromium.org, dsinclair@chromium.org
Link to the patchset: https://codereview.chromium.org/1295753002/#ps20001
(title: "Resolve nits")

[commit-bot: I haz the power, 2015-08-14 15:55:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1295753002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1295753002/20001

[commit-bot: I haz the power, 2015-08-14 16:38:30]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-08-14 16:39:25]

Patchset 2 (id:??) landed as
https://crrev.com/a4c2539f13510a650d51e65204ea52426979659c
Cr-Commit-Position: refs/heads/master@{#343408}

[Mike Lerman, 2015-08-14 19:04:54]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/1294833004/ by mlerman@chromium.org.

The reason for reverting is: This CL flakily causes failures on try bots,
specifically in SingleProcessMemoryTracingTest.BrowserInitiatedSingleDump:

c:\b\build\slave\win_builder__dbg_\build\src\content\browser\tracing\memory_tracing_browsertest.cc(142):
error: Mock function called more times than expected - returning default value.
    Function call: OnMemoryDump(@3286BA0C 4-byte object \u003C01-00 00-00>,
3286B920)
          Returns: false
         Expected: to be called once
           Actual: called twice - over-saturated and active

On Win7 -
https://build.chromium.org/p/chromium.win/builders/Win7%20Tests%20%28dbg%29%2...

and ../../content/browser/tracing/memory_tracing_browsertest.cc:148: Failure
Value of: last_callback_success_
  Actual: false
Expected: true

on Mac 10.8:
https://build.chromium.org/p/chromium.mac/builders/Mac10.8%20Tests/builds/6628
.